© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13544
    Prev
    Next
    INDEX
    Explanations

    error handling, comparisons, counting

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires most strongly on identifiers (including macros) appearing on the left‐hand side of an assignment or definition.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Heilige
    -1.03
    recently
    -1.02
     heilige
    -0.98
     reciente
    -0.96
    previously
    -0.91
     mitään
    -0.91
     являются
    -0.91
     alcune
    -0.90
    zzleHttp
    -0.90
     могут
    -0.88
    POSITIVE LOGITS
     there
    1.20
     but
    0.93
    ضر
    0.88
    ∃
    0.88
     someone
    0.87
     Olen
    0.86
     There
    0.85
     came
    0.85
     प्रतिशत
    0.85
     an
    0.85
    Activations Density 0.004%

    No Known Activations